You know what, I'll actually worry when HP "fires"/parts amicably with guys like Bdale Garbee, Joey Hess, and Matt Taggart that are actually doing real work on Linux and free software.
The bottom line is that Bruce was there for PR purposes, kinda like ESR's job at VA. PR is nice, and the world these days doesn't go around without it (despite the protestations of some of my fellow Debian developers), but I'll take guys working on code to guys they trott out for interviews any day.
I would just say why are they releasing more 2.2 stuff when they should just finalize Woody.
That assumes that the same people that are finalizing woody are also the same people who released 2.2r7. I believe this is a false assumption, although the exact responsibilities of the members of the Cabal (TINC) are kept secret from us mere mortal developers.
Well, except for the minor details that the URL and email address on my post are both valid, and I've got a +1 bonus, I suppose you're right... off to do some goat fucking.
There's a reason they're [OpenBSD] going on 5 years without a remote hack.
Funny, wasn't OpenBSD about the only Unix that turned out to be vulnerable to the OpenSSH hole in its "out of the box" configuration?
Then again, I guess being an AC means you get to make shit up with no accountability...
When Walmart moves into a town they offer very low prices sometimes even lower than cost. After the smaller competitors are forced out of business they return to their regular prices. They may call it Grand Opening prices or whatever. This is illegal in many jurisdictions and unethetical in other areas.
Do you have any evidence of this whatsoever? Wal-Mart sets prices on a regional or nationwide basis; if I walk into the Wal-Mart in Oxford, Miss., everything in stock will be the same price as it is at the Wal-Mart in Tupelo or Batesville or the umpteen stores around Memphis. The only difference is that the Oxford store, a pre-supercenter, doesn't have as good a selection as the others. (At least, this is the case for most products; I don't know about groceries, but that's a very different market than general merchandise.)
The problem is the 2+ year release schedule. I have no problem with not having KDE3 next month. But I don't want to wait 2+ years for it. And on a corporate computer upgrading to testing isn't an option. And wouldn't it be neat to have them using Debian instead of RedHat?
I'd be very surprised if woody+1 takes two years to become stable (as it was, potato was released Monday, 14 August 2000, which was 624 days or 1.7 years ago). Of course, that largely depends on the release manager's goals for woody+1.
From discussion with Debain users (and time spent administering Debian boxed at my workplace) Debian's rpm support doesn't work that well for anything apart from large self-contained statically compiled packages. The problem being that the Linux Standards Base will probably be considered the definition of what a Linux distro is in a couple of years (and is starting to be used as a yardstick these days). Yes Debian ostensibly supports the LSB via alien, but how well?
That's a very good question (and one nobody bothered to answer in the ensuing discussion). At present, it probably doesn't support it at all, excluding the rudimentary support for LSB packages joeyh added to alien 8.00. I can't locate any intent to package statement for lsb or lsb-rpm (the former being the LSB's core dependency, the latter being the RPM specified by LSB). The only packaged LSB component is lsb_release.
I'd have to do a bit more digging to figure out if anyone is actually working on this stuff.
They seem to be focusing on staying two steps behind Microsoft (Evolution, Mono, Gnumeric...)
Slam me if you like, but Gnumeric means I can keep my gradebook and track my business P/L on a free software application instead of dual-booting Windows. People need good spreadsheets and groupware to do real work.
Oh, and "apt-get install evolution-ssl" works just fine here... installing good software for Linux doesn't have to be hard, you know.
I guess where I come down is to argue that if we're going to have a settlement (and since MS has consented to this "Soviet-style technical committee" in lieu of taking its chances at trial), it ought to be a worthwhile one. I would also argue that MS has committed real and serious crimes, including perjury before a federal court, that ought to be illegal regardless of MS's status as a monopoly or not.
FWIW, it seems that most on/. would fail to be satisfied by anything short of the forcible redistribution of Bill Gates' wealth to sweatshop workers in Malaysia and the public execution of Steve Ballmer.
I am writing in opposition to the proposed settlement of the U.S. case
against Microsoft Corporation. In particular, I request that the
following, or something substantially similar, be made a condition of
any settlement with Microsoft:
Microsoft shall be required to disclose all application programming
interfaces (APIs) that are used by any software sold or given away by
Microsoft that is not included with Windows 98, Windows 2000
Professional, Windows XP Home, Windows XP Professional, Windows CE,
Windows XP Tablet PC Edition, or any subsequent Microsoft operating
system implementing portions of the Win32 API, as well as any
Microsoft Middleware component that is offered for free download by
Microsoft, including Internet Explorer ("add-on code"). Microsoft
shall further be required to provide public documentation of these
application programming interfaces, available to software developers
at a reasonable cost, at a similar level of detail to Microsoft's
existing documentation of public APIs, not to exceed the per-page cost
of said documentation. Microsoft shall further be obligated to fully
disclose the effects of API calls made by its add-on code, including
the side-effects of specifying particular arguments to these API
calls.
The technical committee or any third party shall have the authority to
inspect Microsoft's add-on code, in binary form, and documentation to
verify compliance with this provision.
Nothing in this section shall compel Microsoft to provide any such
add-on code for free for use with non-Microsoft operating systems;
however, no add-on code sold for any monetary cost (including
"Microsoft Office") shall require the use of a Microsoft operating
system for execution of said code. Further, nothing in this section
shall limit the technical committee's rights to examine source code as
set forth in the agreement.
I also fully support the position of Dan Kegel, et al, in their
forthcoming letter regarding the settlement.
Christopher N. Lawrence
Ph.D. Candidate and System Administrator
2000 Libertarian candidate for U.S. Congress, 1st District of Mississippi
Oxford, Mississippi
Guido couldn't release it under the GPL, because it wasn't entirely "his" software to license.
More correctly, Guido didn't want to license Python under the GPL, but did want it to be able to be integrated with GPLed software, as well as software under virtually any sort of license.
(25 pages of Python license history snipped... see the full scoop for the current license.)
But then again, the charge that Microsoft used their monopoly to overcharge customers is a fairly weak one. Even if it did hold up in court it would probably result in $600 million dollars spread across 60 million claimants. In other words, the only people who profit are the lawyers.
Indeed, the price Microsoft charges for Windows is well below what any sane economist would describe as the market-clearing price for it, particularly at the bulk OEM level. My recollection is that the market price for Windows would be in the $500-600 range if MS engaged in this pricing package.
I guess you could argue that what MS charges for Windows is actually below cost ("dumping"), but then again you could probably levy the same charge against any OS vendor. But that's only illegal if you're doing it to drive another business out of the market...
(/me kisses his karma goodbye for saying something non-negative about Microsoft.)
It's actually easier to run it over FireWire than over IDE, because the FireWire sbp2 driver exposes CD devices directly to the sg driver. No need for ide-scsi or its analog.
However, your kernel will blow up if you use a stock kernel's 1394 driver to write CDs in "real" mode. Get the CVS code from the linux1394 project site and use either the current or "last stable" branch.
I've been doing some work with 1394 devices under Linux, both as a personal hobby and for my employers. This is what I've been able to determine:
Stability: 1394 storage is pretty stable when you only have one drive on the bus. Multiple drives may make things flaky, particularly when you have lots of IO going on and are using broken drivers.
Speed: performance isn't bad with one drive, but multiple drives are slow. This is mainly due to the use of serialized IO; nonserialized IO is faster but makes things very unstable.
Hotplug: Hotplugging really isn't there yet. You may have to connect and disconnect a device a few times for the 1394 code to recognize it. Once you connect it successfully, you have to run rescan-scsi-bus.sh to get it to show up in/proc/scsi/scsi. Then you can mount it. Unplugging is slightly less hassle: umount, disconnect the device, and run rescan-scsi-bus.sh. The dynamic nature of the bus makes it hard to have a decent fstab with multiple drives; you may want to use volume labels to get around this problem.
Power: all of the units I've seen are self-powered, not bus-powered, so the power isn't a problem.
Cards: most OHCI cards should work with no hassles. I bought the cheapest (~$35) 1394 cards I could find on buy.com and they work just fine (they have a VIA chipset).
My best advice would be to surf over to the Linux1394 project website and read the docs over there; you'll probably want to get their drivers anyway, instead of using what's in the stock kernel.
What if I told you that the UK has TV channels with NO COMMERCIALS! I'm kidding, right? No, there really is!
And guess what, there's NO LOGO, either! (OK, so they've started putting one in for a second or two at the start and end of the program) Is this broadcaster crazy? How does it get its funding?
Now, I know that THE MARKET must dictate everything, and socialism is an EVIL THING that has NEVER WORKED, but guess what, the people of the UK actually collectively pay for these TV channels! And they like that!
And the funny thing is, instead of socialism, the people of the UK collectively pay for it through a mandatory user fee that you have to pay whether or not you watch the programming if you use a television for any purpose in your household. IIRC the commercial broadcasters also have to pay a fee out of their advertising to help support the BBC.
The U.S. also has channels without commercials. The difference is that the people who want those channels pay for them, and the people who don't don't have to. (I exclude PBS, which now has inter-programming commercials, er, I mean "sponsorship announcements.")
Oh, and socialism works great as long as everyone in the system cooperates. The moment free riders show up, however, it becomes a real pain in the ass...
Jim posted details of how he mastered the DVD-Rs under Linux on the debian-cd; however, the interesting part is:
Burning the DVD-R is another story. schilling@fokus.gmd.de is
making noises about making cdrecord (pro) a "for-profit" piece
of software, and has not released the code that
will actually burn a DVD-R......but that is another story.
Oh, very simple, actually. Go re-read the remedy suggested by Judge Jackson. The break-up was only part of it. The most important part, wildly underlooked, was that all technical communications between the different parts of Microsoft would be made public . In short, there would be a Microsoft-OS part that would make the core OS, and the rest of Microsoft, that does IE and the Media Player and everything, couldn't commingle its proprietary apps into the OS without the very way they are commingled becoming public.
This remedy doesn't require two different companies; all it requires is:
All OS APIs must be documented.
All applications must call documented APIs and be developed "clean room" (no access to OS source code that isn't available to competitors).
This solution would be easily enforcible (you compare the DLL accesses to the open API list, and if they call any closed API, they are busted) and doesn't require any break-up. In any event, a break-up would leave X number of companies, still controlled by the same people, so it wouldn't accomplish anything except imposing a lot of restructuring costs on M$.
AFAIK, you can pull the drive out, dd the backup partitions over the main partitions, plug it back in, and you'll have whatever version the unit shipped with. You'll need one of the hack boot disks for your PC to do this. And you'll actually have to read the Hack FAQ to know which partitions to do, because I ain't gonna post step-by-step instructions.
Then disconnect the damn phone line!:)
(Don't blame me if this blows up in your face. I've never tested it! But it should work, unless the 2.0.1 upgrade zapped the backup partitions too. You will probably lose all of your recordings though...)
And tell me how is this supposed to make your television viewing better, huh? This will just result in TV program being adjusted to the lowest common denominator (not that it's not already) and quality programs that are targeted at a bit more discerning viewers will disappear altogether.
Well, for one thing it will help networks determine the viewing habits of people who timeshift TV. Neilsen can't do that reliably (only with the diaries). What does that mean in practice? It means that people who actually are discerning viewers will get more weight in the process, and quality shows will look better than just the crap that happens to be on after Friends this month.
I own 2 TiVos. God knows I can't make any money selling my viewing info (who the hell cares what I watch on TV, anyway?), but if TiVo can do it by aggregating my data with that of a quarter million other people, more power to them!
Specifically, CNRI wants to resolve all disputes in Virginia, which is where they are headquartered and don't have to send lawyers off to New York or California or East Buttwad every time they get sued.
Think about it.
(To put it another way, what part of UTICA would actually benefit CNRI in enforcing a license for free software? Especially when about the only way you could possibly violate that license is to misuse CNRI's name...)
Slashdot Mirror
Uh, yeah, you can get USENET through BellSouth; try using news1.lig.bellsouth.net as your server.
You know what, I'll actually worry when HP "fires"/parts amicably with guys like Bdale Garbee, Joey Hess, and Matt Taggart that are actually doing real work on Linux and free software.
The bottom line is that Bruce was there for PR purposes, kinda like ESR's job at VA. PR is nice, and the world these days doesn't go around without it (despite the protestations of some of my fellow Debian developers), but I'll take guys working on code to guys they trott out for interviews any day.
Well, except for the minor details that the URL and email address on my post are both valid, and I've got a +1 bonus, I suppose you're right... off to do some goat fucking.
No doubt the comment is the result of visiting (or not visiting) the absolute craphole of a Wal-Mart in Ypsilanti.
OTOH, I'd put my Wal-Mart Supercenter up against your Meijer any day...
I didn't realize Adrian Bunk had decided to "participate" in the KDE release process too...
</sarcasm>
From discussion with Debain users (and time spent administering Debian boxed at my workplace) Debian's rpm support doesn't work that well for anything apart from large self-contained statically compiled packages. The problem being that the Linux Standards Base will probably be considered the definition of what a Linux distro is in a couple of years (and is starting to be used as a yardstick these days). Yes Debian ostensibly supports the LSB via alien, but how well?
That's a very good question (and one nobody bothered to answer in the ensuing discussion). At present, it probably doesn't support it at all, excluding the rudimentary support for LSB packages joeyh added to alien 8.00. I can't locate any intent to package statement for lsb or lsb-rpm (the former being the LSB's core dependency, the latter being the RPM specified by LSB). The only packaged LSB component is lsb_release.
I'd have to do a bit more digging to figure out if anyone is actually working on this stuff.
They seem to be focusing on staying two steps behind Microsoft (Evolution, Mono, Gnumeric...)
Slam me if you like, but Gnumeric means I can keep my gradebook and track my business P/L on a free software application instead of dual-booting Windows. People need good spreadsheets and groupware to do real work.
Oh, and "apt-get install evolution-ssl" works just fine here... installing good software for Linux doesn't have to be hard, you know.
Good question. In principle, Libertarians (big and small L) do oppose most government regulation.
/. would fail to be satisfied by anything short of the forcible redistribution of Bill Gates' wealth to sweatshop workers in Malaysia and the public execution of Steve Ballmer.
Maybe a place to start thinking more about this is ESR's essay "Why Libertarians Should Not Love Bill Gates".
I guess where I come down is to argue that if we're going to have a settlement (and since MS has consented to this "Soviet-style technical committee" in lieu of taking its chances at trial), it ought to be a worthwhile one. I would also argue that MS has committed real and serious crimes, including perjury before a federal court, that ought to be illegal regardless of MS's status as a monopoly or not.
FWIW, it seems that most on
Microsoft shall be required to disclose all application programming interfaces (APIs) that are used by any software sold or given away by Microsoft that is not included with Windows 98, Windows 2000 Professional, Windows XP Home, Windows XP Professional, Windows CE, Windows XP Tablet PC Edition, or any subsequent Microsoft operating system implementing portions of the Win32 API, as well as any Microsoft Middleware component that is offered for free download by Microsoft, including Internet Explorer ("add-on code"). Microsoft shall further be required to provide public documentation of these application programming interfaces, available to software developers at a reasonable cost, at a similar level of detail to Microsoft's existing documentation of public APIs, not to exceed the per-page cost of said documentation. Microsoft shall further be obligated to fully disclose the effects of API calls made by its add-on code, including the side-effects of specifying particular arguments to these API calls.
The technical committee or any third party shall have the authority to inspect Microsoft's add-on code, in binary form, and documentation to verify compliance with this provision.
Nothing in this section shall compel Microsoft to provide any such add-on code for free for use with non-Microsoft operating systems; however, no add-on code sold for any monetary cost (including "Microsoft Office") shall require the use of a Microsoft operating system for execution of said code. Further, nothing in this section shall limit the technical committee's rights to examine source code as set forth in the agreement.
I also fully support the position of Dan Kegel, et al, in their forthcoming letter regarding the settlement.
Christopher N. Lawrence
Ph.D. Candidate and System Administrator
2000 Libertarian candidate for U.S. Congress, 1st District of Mississippi
Oxford, Mississippi
More correctly, Guido didn't want to license Python under the GPL, but did want it to be able to be integrated with GPLed software, as well as software under virtually any sort of license.
(25 pages of Python license history snipped... see the full scoop for the current license.)
But then again, the charge that Microsoft used their monopoly to overcharge customers is a fairly weak one. Even if it did hold up in court it would probably result in $600 million dollars spread across 60 million claimants. In other words, the only people who profit are the lawyers.
Indeed, the price Microsoft charges for Windows is well below what any sane economist would describe as the market-clearing price for it, particularly at the bulk OEM level. My recollection is that the market price for Windows would be in the $500-600 range if MS engaged in this pricing package.
I guess you could argue that what MS charges for Windows is actually below cost ("dumping"), but then again you could probably levy the same charge against any OS vendor. But that's only illegal if you're doing it to drive another business out of the market...
(/me kisses his karma goodbye for saying something non-negative about Microsoft.)
It's actually easier to run it over FireWire than over IDE, because the FireWire sbp2 driver exposes CD devices directly to the sg driver. No need for ide-scsi or its analog.
However, your kernel will blow up if you use a stock kernel's 1394 driver to write CDs in "real" mode. Get the CVS code from the linux1394 project site and use either the current or "last stable" branch.
- Stability: 1394 storage is pretty stable when you only have one drive on the bus. Multiple drives may make things flaky, particularly when you have lots of IO going on and are using broken drivers.
- Speed: performance isn't bad with one drive, but multiple drives are slow. This is mainly due to the use of serialized IO; nonserialized IO is faster but makes things very unstable.
- Hotplug: Hotplugging really isn't there yet. You may have to connect and disconnect a device a few times for the 1394 code to recognize it. Once you connect it successfully, you have to run rescan-scsi-bus.sh to get it to show up in
/proc/scsi/scsi. Then you can mount it. Unplugging is slightly less hassle: umount, disconnect the device, and run rescan-scsi-bus.sh. The dynamic nature of the bus makes it hard to have a decent fstab with multiple drives; you may want to use volume labels to get around this problem.
- Power: all of the units I've seen are self-powered, not bus-powered, so the power isn't a problem.
- Cards: most OHCI cards should work with no hassles. I bought the cheapest (~$35) 1394 cards I could find on buy.com and they work just fine (they have a VIA chipset).
My best advice would be to surf over to the Linux1394 project website and read the docs over there; you'll probably want to get their drivers anyway, instead of using what's in the stock kernel.And the funny thing is, instead of socialism, the people of the UK collectively pay for it through a mandatory user fee that you have to pay whether or not you watch the programming if you use a television for any purpose in your household. IIRC the commercial broadcasters also have to pay a fee out of their advertising to help support the BBC.
The U.S. also has channels without commercials. The difference is that the people who want those channels pay for them, and the people who don't don't have to. (I exclude PBS, which now has inter-programming commercials, er, I mean "sponsorship announcements.")
Oh, and socialism works great as long as everyone in the system cooperates. The moment free riders show up, however, it becomes a real pain in the ass...
Can anyone else confirm this information?
- All OS APIs must be documented.
- All applications must call documented APIs and be developed "clean room" (no access to OS source code that isn't available to competitors).
This solution would be easily enforcible (you compare the DLL accesses to the open API list, and if they call any closed API, they are busted) and doesn't require any break-up. In any event, a break-up would leave X number of companies, still controlled by the same people, so it wouldn't accomplish anything except imposing a lot of restructuring costs on M$.I believe George Carlin said it best: the planet's doing fine, it's the people who are going to kill themselves off.
AFAIK, you can pull the drive out, dd the backup partitions over the main partitions, plug it back in, and you'll have whatever version the unit shipped with. You'll need one of the hack boot disks for your PC to do this. And you'll actually have to read the Hack FAQ to know which partitions to do, because I ain't gonna post step-by-step instructions.
:)
Then disconnect the damn phone line!
(Don't blame me if this blows up in your face. I've never tested it! But it should work, unless the 2.0.1 upgrade zapped the backup partitions too. You will probably lose all of your recordings though...)
And tell me how is this supposed to make your television viewing better, huh? This will just result in TV program being adjusted to the lowest common denominator (not that it's not already) and quality programs that are targeted at a bit more discerning viewers will disappear altogether.
Well, for one thing it will help networks determine the viewing habits of people who timeshift TV. Neilsen can't do that reliably (only with the diaries). What does that mean in practice? It means that people who actually are discerning viewers will get more weight in the process, and quality shows will look better than just the crap that happens to be on after Friends this month.
I own 2 TiVos. God knows I can't make any money selling my viewing info (who the hell cares what I watch on TV, anyway?), but if TiVo can do it by aggregating my data with that of a quarter million other people, more power to them!
Specifically, CNRI wants to resolve all disputes in Virginia, which is where they are headquartered and don't have to send lawyers off to New York or California or East Buttwad every time they get sued.
Think about it.
(To put it another way, what part of UTICA would actually benefit CNRI in enforcing a license for free software? Especially when about the only way you could possibly violate that license is to misuse CNRI's name...)